Author: Ada, Deep Tide TechFlow
In the early hours of February 28, the United States and Israel launched a joint military strike against Iran.
The textbook says: When war comes, buy gold.
But this time, the textbook seems to be wrong.
Gold briefly surged from $5,296 to $5,423, then falling all the way to around $5,020, closing lower for two consecutive weeks. Bitcoin rebounded from a panic low of $63,000 to $75,000, rising over 20%, outperforming gold, the S&P, and the Nasdaq.
The same war, the same period. Gold fell, Bitcoin rose.
What exactly happened?
Gold: Strangled by Interest Rates
Gold's performance on the day the war broke out was still normal. On the 28th, the gold price surged 2%, breaking through $5,300. Panic buying poured in, everything looked exactly like the historical script.
Then the script fell apart.
On March 3, the gold price plummeted over 6%, falling to $5,085. It then fluctuated repeatedly between $5,050 and $5,200 for the next two weeks, with no clear direction. At the time of writing, spot gold is around $5,020, nearly 10% down from the all-time high of $5,416 at the end of January.
The war is still going on, shells are still flying, yet gold keeps falling further.
The chain of events is as follows: In this war, the Strait of Hormuz was blockaded. About one-fifth of the world's seaborne oil passes through this waterway. Iran blockaded the strait, insurers withdrew vessel coverage, oil tankers stopped sailing, and oil prices broke through $100. The International Energy Agency urgently released 400 million barrels of strategic petroleum reserves, double the amount during the 2022 Russia-Ukraine war. Daniel Ghali, a commodity strategist at TD Securities, said: "A gap this big cannot be plugged."
Soaring oil prices ignited inflation expectations. The market began repricing the Federal Reserve's interest rate cut path. Before the war, the market still expected two rate cuts in 2026. But according to Bloomberg, traders now see almost zero probability of a rate cut at this week's Fed meeting.
High interest rates are the natural enemy of gold. Gold yields no interest; the higher the interest rates, the greater the opportunity cost of holding gold. Funds naturally flow to interest-bearing assets like U.S. Treasuries. Barbara Lambrecht, a commodity analyst at Commerzbank, pointed out: "The gold price has consistently failed to benefit from this geopolitical crisis. Oil and gas prices rose sharply again this week, increasing inflation risks, which may force central banks to take countermeasures."
The traditional logic is that war causes panic, and panic pushes up gold. But the chain this time changed—war led to soaring oil prices, which triggered inflation, inflation locked in interest rates, and interest rates suppressed gold. What gold fears is not the war itself, but the inflationary consequences brought by the war.
There is another more alarming signal. The head of the Polish Central Bank recently stated publicly that they are considering selling part of their gold reserves to lock in profits. Over the past three years, global central bank gold buying has been the biggest driver of the rising gold price. If even central banks begin to loosen, long-term support for the gold price will develop cracks. Philip Newman, Director of London-based precious metals consultancy Metals Focus, said: "Some investors are disappointed by gold's muted reaction after the outbreak of war and have started to reduce positions. This selling itself, in turn, reinforces the price softness."
Bitcoin: Rising Against the Trend
On February 28, news broke of the US-Israel joint strike on Iran. Bitcoin was the only liquid asset still trading that day; it plummeted 8.5% in minutes, from $66,000 to $63,000.
Gold rose, the dollar rose, Bitcoin fell. Everyone's first reaction was the same: Bitcoin is a risk asset, not a safe-haven asset.
Looking back two weeks later, the situation is more complex than that judgment.
On March 5, Bitcoin rebounded to $73,156. On March 13, it briefly broke through $74,000. At the time of writing, Bitcoin is trading at $73,170, up about 20% from the pre-war low. Over the same period, gold fell about 3.5%, and the S&P 500 fell about 1%.
Bitcoin outperformed all traditional safe-haven assets. This is a fact. But why?
The most popular explanation in the market is: The war leads to fiscal expansion and economic recession, forcing the Fed to eventually cut rates and print money, and loose monetary policy benefits Bitcoin. This narrative sounds compelling but has an obvious logical flaw—if the inflation caused by the war prevents the Fed from cutting rates, "money printing" won't happen. Moreover, even if the Fed does print money, gold also benefits. The simple expectation of "money printing" alone cannot explain the divergence between gold and Bitcoin.
A more honest answer is the叠加 (superposition) of several factors.
First, a oversold technical rebound. Bitcoin fell from its all-time high of $126,000 in October last year to $63,000, a drop of about 50%. In early February, a sudden wave of liquidations wiped out $2.5 billion in leveraged positions over a weekend. Analysis from CoinDesk suggests this liquidation "cleared out the weakest holders, reset market positions," leaving a leaner market. So when the war came, Bitcoin didn't have much weak hands left to be sold off panic.
Second, the structural advantage of 24/7 trading. February 28 was a Saturday. When the US and Israel struck Iran, global stock, bond, and commodity markets were closed. Bitcoin was the only open liquidity window. It got sold off first because panic-stricken money needed instant liquidation; but it was also the only place to承接 (receive) returning funds before markets opened on Monday.
Third, ETF fund flows. U.S. spot Bitcoin ETFs saw net inflows of over $1.34 billion in March, with three consecutive weeks of net inflows, the longest streak since July last year. BlackRock's IBIT alone attracted nearly $1 billion in new money in March. Meanwhile, the world's largest gold ETF (SPDR Gold ETF) saw outflows of over $4.8 billion during the same period. Money is moving, but this looks more like institutions reallocating portfolios. It's too early to conclude if this constitutes a long-term trend.
Fourth, portability during war. This factor is rarely mentioned in mainstream analysis but is extremely important in the specific context of the Middle East war. Dubai is a core hub for global gold trading, connecting European, African, and Asian markets. After the war broke out, Dubai's gold logistics network was severely impacted, routes were interrupted, insurance became invalid, and physical gold was trapped in warehouses unable to be shipped out. You can't carry a ton of gold bars through a war zone. Bitcoin is the complete opposite—a person can carry nothing, remember 12 seed words, cross the border, and effectively take all their wealth with them. After the war broke out, outflows from Iran's largest crypto exchange, Nobitex, surged 700%. This isn't investors being bullish on Bitcoin; this is people voting with their feet in a war, choosing the easiest thing to take away.
Tiger Research noted in a report: "In finance, a 'safe haven' refers to an asset whose price remains stable during a crisis. This is a completely different concept from 'an asset that can be used during a crisis.'" Bitcoin in this war clearly belongs to the latter category.
No single factor can explain everything alone. But together, they can explain why Bitcoin performed better in this war than most people expected.
Two Surprises
Putting these two narratives together, this war created two surprises.
The first surprise was gold. It fell when it should have risen the most. This war directly hit energy supply, triggering not simple panic but inflation. Inflation expectations suppressed the gold price through the interest rate channel. Gold's safe-haven function is not unconditional—when the transmission path of war is crisis leading to inflation and interest rates cannot fall, gold gets stuck in the middle, unable to move. There's also a often overlooked physical weakness: in war, physical gold is hard to move.
The second surprise was Bitcoin. It rose when it should have fallen the most. But this does not mean Bitcoin has "matured" into a safe-haven asset. Its performance is more like the叠加 (superposition) of multiple technical factors and structural advantages. Aurelie Barthere, Chief Research Analyst at Nansen, noted that Bitcoin's downside sensitivity to war news has significantly decreased, and the European Stoxx index fell harder than Bitcoin during the same period. CoinDesk's analysis is more accurate: "Bitcoin is not a safe haven, nor is it a pure risk asset. It has become a 24/7 liquidity pool that absorbs shocks when other markets are closed, faster than anything else."
With every piece of war escalation news, Bitcoin still falls. It just falls less each time and bounces back faster.
Old Map, New World
Over the past five years, the market told a simple and powerful story: Gold is the anchor in troubled times, Bitcoin is digital gold.
The March 2026 Middle East war tore this story apart.
Gold's millennia-old safe-haven credit hasn't collapsed, but it exposed a weakness rarely clearly written in textbooks: When the transmission path of war is inflation rather than simple panic, interest rates are more powerful than geopolitics. Bitcoin outperformed gold, but this does not mean it has taken over the banner of "safe-haven asset." Its rise is the result of the simultaneous发力 (force application) of four lines: oversold rebound, structural advantage, institutional allocation, and war portability—not the market's formal coronation of its identity.
Subsequent trends depend on two variables: how long this war lasts, and what the Fed ultimately chooses. Gold and Bitcoin are betting on different outcomes of the same war, and the outcome is not yet clear.
The term "safe haven" might need redefinition after this war. It is no longer a label for an asset class, but a question about the time dimension: are you hedging against today's risk, or betting on tomorrow's world?
Gold and Bitcoin gave two completely different answers.
